You are on page 1of 13

Available online at www.sciencedirect.

com

ScienceDirect
Procedia Engineering 66 (2013) 451 463

5th Fatigue Design Conference, Fatigue Design 2013

Pile fatigue assessment during driving


Jean Chung, Rgis Wallerand, Morgane Hlias-Brault
Subsea 7, 1 Quai Marcel Dassault, 92156 Suresnes Cedex, France

Abstract
The aim of this paper is to present the methodology currently followed in the offshore industry for the prediction of the pile
fatigue induced by pile hammering during installation. A few decades ago piles were thicker, of lower yield stress and hammers
were less efficient. Today, to be cost effective, it is common to use high strength steel resulting in lower pile thickness and
increased driving stress together with better hammer efficiency to reduce installation time. Hence fatigue damages are more
important. Evaluation of fatigue damages depend on various parameters such as applied stresses, number of blows, Stress
Concentration Factor (SCF) and S-N curves. It is shown how fatigue damages are dependant on the choice of the values used for
the SCF and S-N curves and also on applied stresses by the hammer, and in the end the impact of these parameters on the residual
available fatigue life for the in-place conditions.

byby
Elsevier
Ltd.
2013
2013The
TheAuthors.
Authors.Published
Published
Elsevier
Ltd. Open access under CC BY-NC-ND license.
Selection
responsibility
of CETIM,
Direction de l'Agence de Programme.
Selectionand
andpeer-review
peer-reviewunder
under
responsibility
of CETIM
Pile; Driving; Fatigue

1. Introduction
Offshore platforms are subjected to harsh environment and variable loads such as waves, currents and winds.
When designing a platform supported by a substructure called jacket, fatigue analysis is performed for all tubular
nodes and the foundation piles for in-place conditions.
The common way to install the piles is by driving with an impact hammer. The piles are generally subjected to
high dynamic loads close to the steel yield stress. The pile fatigue damage is defined during driving for all the butt
welds of the pile either done offshore for the assembly of pile sections or in the yard for the prefabrication of the pile
sections.
The available fatigue life for in-place conditions is thus reduced due to initial fatigue damages for pile installation
during driving. Piles welds cannot be inspected and/or repaired as they are embedded in soil or inserted in jacket

1877-7058 2013 The Authors. Published by Elsevier Ltd. Open access under CC BY-NC-ND license.
Selection and peer-review under responsibility of CETIM
doi:10.1016/j.proeng.2013.12.098

452

Jean Chung et al. / Procedia Engineering 66 (2013) 451 463

legs. Hence safety factor for fatigue damages requested by Total, Shell or Exxon Mobil is equal to 10. Usual service
life of an offshore structure is 20 years which means a fatigue life of more than 200 years. Moreover, in some cases
the structures are built in yards which are located very far from the field with transport duration about a month.
Hence fatigue damages due to transportation loads have also to be accounted for in the determination of the residual
fatigue life for in-place conditions.
The input to the fatigue analysis comes mainly from the pile driving assessment. A description of the most
common method followed to assess the Soil Resistance to Driving (SRD), and the way the pile driveability is
derived will be made. The pile driveability provides the range of stresses experienced by the piles during the driving
process and the number of impacts required to drive the piles to the required penetration. A good pile driveability
analysis is thus crucial to a good fatigue assessment, and data from the field are most wanted to guarantee improved
predictions.
2. Characteristics of offshore piles
The most common type of offshore pile is a pipe pile, which is typically driven into the seabed by a hammer.
Common pile outer diameter (OD) ranges from 30" for tripods to 84" for big 8 legs platforms. Pile wall thickness
starts from 1". Embedded length in seabed ranges from 80m to 100m for Gulf of Guinea platforms for instance. Pile
make-up consists of the detailed design of wall thickness and other features of pile sections. There may be a driving
shoe at the lower end, tapered for easier driving, and with a thicker wall section to cater for stress concentrations and
non-uniformities in the soils and rocks encountered. The central sections may have a thinner wall because they will
support less loads during the operational phase of the platform design life. The upper sections and the sections at
mud line level may have a thicker wall because of the larger stresses they see during operations.
3. Pile driving offshore
The principal methods of installing an offshore driven pile are described by Toolan and Fox (1977). Figure 1
illustrates the driving operation for a pile that is installed through a jacket leg. With the jacket supported by mudmats, the first few sections of pile are lowered to the seabed and allowed to break through a seal, if present. A
hammer is installed on the stack, and is used to drive a segment of the pile into the seabed. When one segment has
been driven to the limit of travel, the piling equipment is lifted off and about 1 m of top of pile is cut off. Another
pile segment is lifted on and welded in place. The girth weld is normally subjected to non-destructive testing
(Magnetic Particle Inspection and Ultrasonic Testing), after which the piling equipment is lifted back on. The
operation is repeated until the required pile penetration below the seafloor has been achieved. The hammer is then
removed, and shims are welded in the annulus between the pile and the leg.

Figure 1 Offshore pile installation sequence

Jean Chung et al. / Procedia Engineering 66 (2013) 451 463

4. Effect of a hammer blow


Figure 2 shows typical elements of an offshore hammer. The hammer has a driving system, drop weight or ram,
anvil, one or more cushions, and a temporary pile cap or cage or helmet. The ram maybe guided as a piston in a
chamber, or by a central rod. The working fluid may be air, steam, oil, or a dieselair mixture. The driving system
may be an external generator of fluid under pressure (external combustion systems), or may be integral with the ram
chamber (diesel hammers).

Figure 2 Offshore hammer


In single-acting hammers, high pressure is used to drive the ram upwards in each stroke, with gravity used to
drop the ram onto the anvil. In double-acting hammers, high pressure is used to move the ram in both directions.
The driving system lifts the ram and drops it onto an anvil or striker plate, or onto a cushion or cap block. The
cushion absorbs some of the damaging high-frequency components of the blow, and helps to spread the stress
evenly across the width of the element beneath it.
When the hammer hits the anvil, stress is transmitted through the cushioning systems into the top of the pile. The
top of the pile moves downwards, and a compressive stressstrain wave starts to travel down the pile.
The wave travels at the speed of sound in steel, about 5100 m/s. Energy is transmitted into the ground through
frictional slip at the soilpile interfaces. This causes a shear wave to travel outwards from the pile into the soil. The
consequent loss of energy reduces the energy of the stress wave travelling down the pile, an effect that can be
modeled as radiation damping.
A partial reflection occurs when the wave reaches the seafloor, although this is minor except for a very stiff
seafloor. Partial reflections also occur whenever the stress wave reaches a boundary between soil layers with
different stiffnesses, and at imperfections and changes in the section properties of the pile. This can be advantageous
because equipment can be installed to measure and analyze the reflected wave received at the pile head. This allows
problems to be identified early.
The information can also be used to assess the ultimate axial capacity of the pile (Likins et al., 2008). When the
compressive stress wave reaches the pile tip the pile shoe moves rapidly into the ground, transmitting more energy
into the ground. The ground does not fully spring back: instead, there is some permanent set there. This consumes
some more energy. The remaining energy is then reflected and travels upwards into the pile, sometimes as a tensile

453

454

Jean Chung et al. / Procedia Engineering 66 (2013) 451 463

wave (sign of stress and displacement in opposition with the compression wave) which produces some elastic
rebound in the upper parts of the pile. The overall elastic rebound may be fully dissipated by soilpile friction,
before the wave reaches the pile top.
One result of these events is a permanent set at the top of the pile. The set can be deduced from the blow-count,
typically expressed in blows per foot or quarter of a meter. Easy driving corresponds to blow-counts of around 10
blows/foot. Hard driving is above 50 blows/foot. Refusal is generally defined in the contract, and is taken to have
occurred if the blow-count reaches 200-300 blows/foot.
5. Pile drivability methodology
In a pile drivability study, upper and lower bound predictions are made of the numbers of hammer blows per foot
of penetration needed to drive the pile into the seabed, and of the maximum compressive and tensile stresses
induced in the pile during driving. Predictions depend on the characteristics of the hammer and the associated
equipment, the pile dimensions, the soil properties, and how far the pile has penetrated into the seabed. The
following strategy is usually used:
(a)
(b)

(c)

For a given pile diameter and wall, estimate the upper and lower bounds of the soil resistance to driving.
The SRD is extracted from the ultimate axial pile capacity that would occur under static loading conditions.
The estimates are plotted as a graph of the SRD versus pile tip penetration into the seabed.
For a given pile-driving hammer and required final pile tip penetration, calculate the relation between the
SRD and hammer blows required to drive the pile per foot or per quarter of a meter of movement. This is
done using a wave equation analysis described below. It gives a bearing graph. Upper and lower bound
graphs may be needed, but a single graph can be used if the bounds are close.
For each pile penetration and bound, the SRD is read from the SRD profile, the corresponding blow-count
is determined from the relevant (upper or lower bound) bearing graph, and the result is plotted on the blowcountpenetration graph.

As well as blow-count data, maximum compressive and tensile stresses will be calculated in step (b). The pile
designer will estimate the fatigue damage caused to the pile by combining this information with results from step
(c).
5.1. Calculations for the SRD
The SRD is the ultimate axial pile capacity that is experienced during the dynamic conditions of pile driving.
Predictions of the SRD are usually calculated by modifying the calculation for the ultimate static axial pile capacity
in compression. API RP 2A and ISO 19002 refer to several methods proposed in the literature.
The recommendations of Stevens et al. (1982) are widely used. The recommendations were based on 58 case
histories of installations of large-diameter pipe piles at 15 sites in the Gulf of Mexico. The case histories were
analyzed using a coefficient of lateral earth pressure (ratio of horizontal to vertical normal effective stress) K=0.7 for
sand (API RP 2A now uses 0.8), and with the 1981 API RP 2A method for clay. This is now superseded (Randolph
and Murphy, 1985), but is given in the API commentary. Some alternatives for hard soils are proposed by Colliat et
al. (1993). The approach for carbonate sands is very different now (Kolk, 2000; Rausche and Hussein, 2000).
A designer using Stevens et al. is recommended to start from the 1981 API RP 2A calculation to evaluate the
ultimate axial pile capacity. This code provides guidance to calculate shaft friction f and end bearing q in order to
evaluate the ultimate bearing capacity Qd:
(1)
Qd= Qf + Qp = f As + q Ap
With:
Qf = skin friction resistance,
Qp = total end bearing,
f = unit skin friction capacity (see section 6.4 on pile capacity for axial bearing loads of API RP 2A for
more details on f calculation),

Jean Chung et al. / Procedia Engineering 66 (2013) 451 463

As = side surface area of pile,


q = unit end bearing capacity (see section 6.4 on pile capacity for axial bearing loads of API RP 2A for
more details on q calculation),
Ap = gross end area of pile.
This is then modified, to obtain four curves of the SRD versus depth:
x upper bound, pile assumed plugged,
x upper bound, pile assumed coring,
x lower bound, pile assumed plugged,
x lower bound, pile assumed coring.
A plugged behavior is assumed when the soil plug inside the pile moves with the pile during driving, a coring
behavior is assumed when there is a relative movement between the pile and the soil both on the outside and outside
wall of the pile. These considerations modify the surfaces As and Ap considered in equation (1).
The lower bounds are estimates for the case of continuous driving. The upper bounds may go some way towards
accounting for set-up and for uncertainties in soils data or hammer performance.
The modified curves are determined using soil properties determined from site investigation data in a way that
would give a reasonable upper bound on static capacity, rather than a reasonable lower bound that is used in a
capacity calculation.
5.2. The one-dimensional wave equation
A wave equation analysis is a calculation that takes account of the dynamic response of the pile and soil during
driving.
The relationship between SRD and blow count for a given hammer/pile/soil combination can be derived using
commercially available pile driving software package such as GRLWEAP (PDI, 2010), TNOWAVE, and others,
which uses the wave equation approach prepared by Smith (1960).
In Smith approach a one dimensional formulation for longitudinal wave transmission due to end impact is
followed. To simplify, it is assumed that the input energy of the driving equipment (minus the loss in hammer
mechanism) is equal to pile resistance (including soil tip and side resistance and pile stiffness) multiplied by its
movement through the soil or permanent set. The hammer and pile system is discretized into a number of distinct
elements: masses and springs. Hammer system is considered as masses and springs acting on top of the pile and the
pile is divided into a series of masses connected by weightless springs to represent the pile stiffness. It is found to be
important to model the several distinct components of the hammer system as the ram, cap block, pile cap and
cushion. Springs and dashpots are used to represent the frictional resistance of the soil and the point resistance from
the soil below the pile toe.
Based on the method described above, available software packages can simulate motions and forces in the
foundation pile when driven by an impact hammer and computes the following outputs:
x The blow count of a pile under one or more assumed ultimate resistance values and other dynamic
resistance parameters for soil (depending on the pile penetration), given a particular hammer and driving
system,
x The axial stresses in the pile, averaged over its cross section associated with the assumed capacity value(s),
x The energy transferred by the hammer to the pile for each capacity analyzed.
From these results, the following can be indirectly determined:
x The pile(s) bearing capacity at the time of driving or re-striking, given its observed penetration resistance
(blow count),
x The stresses during driving,
x A suitable hammer for driving a given pile in a given set of soil conditions.

455

456

Jean Chung et al. / Procedia Engineering 66 (2013) 451 463

5.3. Pile driving fatigue


The practice presented here refers to DNV-RP-C203. Justified deviations have been sought and obtained in
recent projects. Fatigue damage due to pile driving is calculated using the methodology presented here above and
extracting the results obtained with GRLWEAP software.
The basic methodology includes:
x Using GRLWEAP outputs, calculate blow counts versus depth and associated stress range,
x Appropriate Stress Concentration Factors (SCF) and S-N curves are selected,
x Using the number of blows and stress range data, driving damage per depth increment is calculated.
Palmgren-Miner rule is then applied to sum the incremental damage and obtain the total damage due to
driving.
Cumulative fatigue damage D of the piles is evaluated with Palmgren-Miner rule:

(2)
 
With:

Ni = number of cycles at failure for a given stress ratio variation i (see section 5.3.2 for more details),
ni = number of cycles inflicted to the structure for the i stress variation.
It should be noted that the stress ratio variation i is not constant with time as during pile driving the energy of
the hammer is adapted to soil resistance.
5.3.1. Stress concentration factor
DNV-RP-C203 provides guidance on fatigue design of offshore steel structures and more precisely on the
calculation of the stress concentration factors to be adopted in fatigue analysis.
For piles with wall thickness transitions, which can be seen as tubular butt weld connections (see figure below),
DNV-RP-C203 (revision 2000 used for the following project described) recommends the SCF to be calculated using
equation (3) below:
1
`e-D
SCF = 1 +^ 6(Gt + Gm )
t
1+(T/t)2.5

(3)

D = 1.82L
1
Dt 1+>T/t@2.5

(4)

Where:

Gm = eccentricity due to concentricity, out of roundness and centre of eccentricity; eccentricity due to out
of roundness giving normally the largest contribution to eccentricity,
Gt = 0.5 (T-t) eccentricity due to a change in thickness,
T = thickness of thicker plate,
D = nominal diameter of tubular connection,
t = thickness of thinner plate,
L= length over which the eccentricity is distributed.

Figure 3 Circumferential butt weld made from one side

Jean Chung et al. / Procedia Engineering 66 (2013) 451 463

5.3.2. S-N Curves


S-N curves are used to determine the fatigue life of a structural steel joint (see figure below). The relevant S-N
curve is a function of the weld type used to join the pile sections.

Figure 4 S-N curves in air extracted from DNV-RP-C203


The form of the S-N curve and the number of cycles at failure for a given stress ratio variation is given by:
 

(5)

Where:
N = predicted number of cycles to failure for stress range ,
= intercept of the design S-N curve with the log N axis,
m = negative inverse slope of the S-N curve,

= the scale effect with:

tref = reference thickness equal 25 mm for welded connections other than tubular joints,
t = pile thickness, t = tref is used for thickness less than tref,
k = thickness exponent on fatigue strength as given in DNV RP-C203.
Values of , k and m are provided in Table 2.3-2 of DNV-RP-C203 for the relevant S-N curve. More details
on the weld type and associated S-N curve are provided in Appendix 1: Classification of structural details of DNVRP-C203 (see extract in table below). It should be noted that for driving analysis an S-N curve F or F3 is usually
considered depending of the presence or not of a backing bar.

457

458

Jean Chung et al. / Procedia Engineering 66 (2013) 451 463

Figure 5 Hollow sections structural details extracted from DNV-RP-C203


6. Pile driving fatigue analysis from a past project
From a past project in the Gulf of Guinea, the following pile driving input data can be summarized:
x Pile is of uniform thickness, that is 42" OD x 2.0" WT with a nominal yield stress of 345 MPa,
x Pile is divided in 4 segments P1 to P4
x Yard welds are made from one side only without a backing element,
x Offshore welds are made from one side only but with a backing element due to the presence of the pile
stabbing guide used as an installation aid for installing the next pile section on the previously installed pile
element.
6.1. Pile driving results
The driving results of the piles down to the target penetration of 95.5m can be summarized in the following table.
Depths

16.9 m to 23 m
44.5 m to 50 m
71.3 m to 73.4 m
73.4 m to 77 m
23 m to 31 m
50 m to 58 m
77 m to 85 m
31 m to 44.5 m
58 m to 71.3 m
85 m to 95.5 m

Pile sections
being driven

Hammer
setting energy
(%)

P1+P2
P1+P2+P3
P1+P2+P3+P4
P1+P2+P3+P4
P1+P2
P1+P2+P3
P1+P2+P3+P4
P1+P2
P1+P2+P3
P1+P2+P3+P4

50

Maximum
compressive
stress in pile
(MPa)
185

Number
of primary
blows
1530

Associated
secondary
stress
(MPa)
219
92

Number of
secondary
occurrences
119
1411

60
75

202
225

1325
3292

88
158
109

1325
189
3103

100

254

4489

53
72

1200
3889

Total number
10636
of blows
Table 1 Summary of results for pile driving fatigue analysis

Jean Chung et al. / Procedia Engineering 66 (2013) 451 463

Figure 6 Example of output of pile stresses


From the above table it can be seen that, for each individual maximum compressive stress the blow wave induces
secondary compressive or tensile stresses. Sometimes the maximum delta stress is even higher than the main
compressive stress but the number of occurrences is low.
6.2. Selection of SCF
Stress Concentration Factors (SCF) to be used have a huge effect on fatigue damages. So, the choice of this factor
should be carefully done.
The value of SCF for butt weld from an important oil Company is given in their specifications and is equal to
1.27 considering a misalignment of 10% of pile wall thickness.
From OTC 7453, a formula is given for the calculation of the SCF at butt weld. In our case, with a constant pile
thickness and considering a misalignment of 10% of pile wall thickness the value of SCF becomes 1.32 which is
slightly greater than the above value of 1.27.
SCF for butt welds can be issued from DNV RP-C203 May 2000 page 18. Considering again a misalignment of
10% of pile wall thickness and a 30-30 bevel weld, SCF is found to be 1.24 which is slightly less than 1.27.
Hence, at that time, SCF were close to each other ranging from 1.24 to 1.32 in our case.
However, in the next revisions of DNV RP-C203 as from October 2001 page 18, it is said that a misalignment of
10% of pile wall thickness is already inherent in the S-N curve, accordingly Gt + Gm is reduced by G0 = 0.1t in

459

460

Jean Chung et al. / Procedia Engineering 66 (2013) 451 463

equation (3). In that case the value of SCF is reduced to 1.0 considering that Gm is equal to 0.1t and Gt is equal to zero
(no thickness variation).
6.3. Pile fatigue results
Damage ratios have been calculated as per HSE, AWS and DNV and are shown thereafter.
6.3.1. HSE
For butt welds from one side only with permanent backing (offshore welding), the F curve is used and is obtained
from P curve in air with a classification factor of 1.34 which is to be applied on the nominal stress.
SCF considered is 1.27 together with a scale effect, (t/t ref)k with tref of 16 mm, with pile wall thickness t of 50.8
mm and with the exponent coefficient k of 0.3. So and No are respectively the stress range and the allowable
number of cycles at the change of slope in the S-N curve. N is the allowable number of cycles at stress range and
n is the applied number of stress cycles.

Figure 7 Fatigue damages as per HSE


The damage is already greater than 1 without any additional Company safety factor. It can also be seen that
damage due to secondary waves is negligible compared to damage caused by primary waves. Hence, we could
neglect damage due to secondary waves in the coming pages. If we consider the case of yard weld without
permanent backing the classification factor increases to 1.52 and, as expected, the damage increases also and
becomes 1.762.
Hence, with HSE rules the damage is high and even greater than 1 due only to pile driving stresses.
6.3.2. AWS D1.1
We use the C1 curve for butt splices, complete joint penetration groove welds, as welded. This curve is issued
from Table 2.6 for circular sections. SCF is again taken as 1.27. There is no thickness correction in the
specifications and influence, if any, of backing bar is not considered.

Jean Chung et al. / Procedia Engineering 66 (2013) 451 463

Figure 8 Fatigue damages as per AWS


Total damage is less than 1 but does not still include Company safety factors. This curve is not much used today
because of the absence of thickness correction and differentiation between with or without permanent backing strip.
6.3.3. DNV RP-C203
This DNV rules are today commonly used for assessing fatigue damages including pile driving damages.
In the revision of May 2000, calculated SCF was equal to 1.27 considering a misalignment of 10% of pile wall
thickness. The SCF reduces to 1.0 in the version from October 2001 as discussed above.
For the case of offshore weld with a permanent backing element, we shall use the F curve for circumferential butt
weld. A scale effect, (t/tref)k with tref of 25 mm and exponent coefficient k of 0.1. A comparison is made with SCF
equals either to 1.27 or 1.0. The case for yard weld without backing strip is also studied.

Figure 9 Fatigue damages as per DNV for two values of SCF and F curve

461

462

Jean Chung et al. / Procedia Engineering 66 (2013) 451 463

Figure 10 Fatigue damages as per DNV for two values of SCF and F3 curve
It can be seen that damage is reduced by more than 2 by considering a SCF of 1.0 instead of 1.27. Hence, DNV is
now less conservative considering the 10% misalignment already included in the S-N curve.
However, using F3 curve has the inconveniency of doubling the damage ratio. For yard welds SCF can also be
considered as equal to 1.0 because the butt weld is performed by automatic machines.
7. Conclusions
It can be seen that evaluating pile driving fatigue damages is not an easy task and depends a lot on the input data.
This is now becoming a bit easier as some Oil Companies now include in their specifications the DNV RP-C203 but
there is still nothing said about what has to be done to evaluate pile fatigue damages.
Nevertheless, the smallest damage ratio of 0.227 using DNV rules is still high and does not comply with safety
factor of 10 as required by some Oil Companies. Discussions with Company to reduce this safety factor may be
necessary, even if the welds are not accessible and cannot be repaired or inspected, the loads during driving are only
temporary and are not present during all field life.
Moreover we shall not forget that, to this pile driving damage, we have to add fatigue damages due to in place
conditions. Fortunately fatigue damages due to in-place conditions can be considered negligible by having the butt
weld far from where the bending moment in the pile is a maximum.
However, according to A.H. Priest and I.M. Gaunt, it appears that pile driving which induces compressive
stresses has no effect on the fatigue life of the butt welds. This could be explained by the initial residual axial
stresses, due to process, in the welds which are tensile on the root side and compressive on the toe side of the welds.
Hence on the root of the weld the additive compressive axial stresses of driving counterbalance the initial residual
axial tensile stresses. On the toe side the axial compressive stresses are increased and yielding occurs. When the
driving compressive loads are removed the final residual compressive stresses decreased due to the yielding of the
steel. And as there is equilibrium between axial compressive and tensile stresses, tensile residual stresses are also
reduced on the root side. This suggests that driving of piles could be a phenomenon that produces stress relief and
would be rather beneficial. In that case there would be no need to account for driving fatigue damages in the
evaluation of pile fatigue for in-place conditions. Of course, this has to be confirmed with large scale tests.

Jean Chung et al. / Procedia Engineering 66 (2013) 451 463

References
[1] A.H. Priest and I.M. Gaunt, The effect of pile driving on the fatigue of welded tubes. Eighth International
Conference on Offshore Mechanics and Arctic Engineering, The Hague, March 19-23, 1989
[2] Barbour, R.J. and Erbrich C.T., 1994. Analysis of in-situ reformation of flattened large diameter foundation piles
using ABAQUS. Proc. UK ABAQUS Users Conf., Oxford.
[3] Aldridge, T.R., Carrington, T.M. and Kee, N.R., 2005. Propagation of pile tip damage during installation. Proc.
Int. Conf. Frontiers in Offshore Geotechnics, eds S. Gourvenec and M. Cassidy, Taylor and Francis, 823827.
[4] Toolan, R.E. and Fox, D.A., 1977. Geotechnical Planning for Piled Foundations for Offshore Platforms. Proc.
ICE London, Part 1, 221243.
[5] Likins, G., Piscsalko, G., Roppel, S. and Rausche, F., 2008. PDA Testing: State of the Art. Proceedings of the
Eighth International Conference on the Application of Stress Wave Theory to Piles, Lisbon, Portugal, 395402
[6] API, 2000. RP2A, Recommended Practice for Planning, Designing and Installing Fixed Offshore Platforms
Working Stress Design, 21st edition, and subsequent updates, American Petroleum Institute.
[7] ISO, 2007. International Standard ISO 19902. Petroleum and natural gas industries: Fixed steel offshore
structures, International Standards Organization.
[8] Stevens, R.S., Wiltsie, E.A. and Turton, T.H., 1982. Evaluating Pile Driveability for Hard Clay, Very Dense
Sand, and Rock. Paper OTC 4205, Offshore Technology Conference.
[9] Randolph, M.F. and Murphy, B.S., 1985. Shaft Capacity of Driven Piles in Clay. Paper 4883, Offshore
Technology Conference.
[10] Colliat, J.L., Vergobbi, P. and Puech, A., 1993. Friction, Degradation and Set-Up in Hard Clays Offshore
Congo and Angola. Paper OTC 7192, Offshore Technology Conference.
[11] Kolk, H.J., 2000. Deep foundations in calcareous sediments. Proc. Int. Conf. Engineering for Calcareous
Sediments, ed. K.A. Al-Shafei, Balkema, 2, 313344.
[12] Rausche, F. and Hussein, M., 2000. Pile driving in calcareous sediments. Proc. Int. Conf. Engineering for
Calcareous Sediments, ed. K.A. Al-Shafei, Balkema, 2, 345359.
[13] PDI, 2010. GRLWEAP (Computer Program for) Wave Equation Analysis of Pile Driving, Version 20101.
Pile Dynamics Inc.
[14] DNV, 2000, Recommended Practice. DNV-RP-C203, Det Norske Veritas
[15] DNV, 2001, Recommended Practice. DNV-RP-C203, Det Norske Veritas
[16] HSE, Offshore Installations : Guidance on design, construction and certification, Fourth Edition, Section 21,
Steel
[17] AWS D1.1:2010, Structural Welding Code-Steel
[18] OTC 7453, Structural Design, Fabrication, and Installation of Offshore Conductor Pipe, G.R. Lang Jr and B.J.
Wood, Mobil Research & Development Corp.

463